The major objective is to define mechanisms by which patterns of visual connections are established during development. This program is a continuation of our efforts to explain how lateral geniculate nucleus (LGN) layers form. We will use the 6 layered tree shrew LGN as a model system since all cell 6 layers develop postnatally. The 3 specific aims examine how retinal afferents influence cell layer formation. In Aim 1, studies are proposed to elucidate the sequence and details of connections of developing LGN cells. Prenatal studies utilizing HRP and 3H amino acid tracers will examine initial retinal afferent segregation. Postnatal studies utilizing HRP will examine the morphology of individual terminal arbors of retinal, tectal, and cortical axons within the nucleus. Also, tracers will be used to examine the sequence of LGN innervation of striate cortex. Aim 2 will provide information on the regulation of cell number and cell death in the retina and LGN during LGN layer formation. Finally, Aim 3 will test different ways in which retinal afferents might trophically influence LGN targets by 1) manipulating optic nerve activity (darkrearing or tetrodotoxin injection), 2) blocking of axoplasmic transport (colchicine), 3) damaging discrete retinal regions. In all cases, LGN cell layer development and cytologic differentiation will be examined to determine the effectiveness of these manipulations. Together these studies should provide a clearer picture of how lamination develops within the LGN; results which in turn could provide clues as to basic rules by which sensory afferents influence the cytologic differentiation and topography of postsynaptic targets in a highly ordered system. As such, these studies are relevant not only to understanding basic mechanisms of normal visual development, but also clinically to understanding disorders that may effect that development.